Governance of fuzzy systems to predict driver perception of service quality

International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
INTERNATIONAL JOURNAL OF ADVANCED RESEARCH IN
6480(Print), ISSN 0976 – 6499(Online) Volume 4, Issue 1, January- February (2013), © IAEME
ENGINEERING AND TECHNOLOGY (IJARET)
ISSN 0976 - 6480 (Print)
ISSN 0976 - 6499 (Online)
IJARET
Volume 4, Issue 1, January- February (2013), pp. 35-41
© IAEME: www.iaeme.com/ijaret.asp ©IAEME
Journal Impact Factor (2012): 2.7078 (Calculated by GISI)
www.jifactor.com

GOVERNANCE OF FUZZY SYSTEMS TO PREDICT DRIVER
PERCEPTION OF SERVICE QUALITY OF VARIABLE MESSAGE
SIGNS
S.G.Uma Sankara, Dr.G.Kalivarathanb
a
Research Scholar, CMJ University, Meghalaya, Shillong.
b
Principal/ PSN Institute of Technology and Science, Tirunelveli, Tamilnadu, Supervisor,
CMJ university, Shillong. Email:sakthi_eswar@yahoo.com

ABSTRACT

Advanced traveler information systems (ATIS) are an intelligent transportation sub
area that has significant interactions between the systems and humans, whether they are
vehicle operators, passengers, or pedestrians. Given this circumstance, consideration of how
drivers perceive and evaluate the service quality provided by these systems is an important
factor in evaluating the performance of these systems. An important element in the
transmission of information to travelers, as part of an ATIS, is the variable message sign
(VMS). In evaluating the service quality by VMS, there are two methods that have been
generally used. One is to evaluate the service quality indirectly through the investigation of
traffic operational effects. For example, how much does the installation of ATIS along the
road increase average vehicle speed or reduce average delay? Investigating these effects is
relatively easy, but it is difficult to truly evaluate how drivers perceive the service quality of
these devices or how satisfied drivers are with the service they are receiving using these
measures of effectiveness (MOEs).

1.0 INTRODUCTION

Another way to investigate driver satisfaction of service quality would be to use a
survey. For instance, motorists could be asked about their satisfaction with the contents and
accuracy of information provided by a VMS. By employing a survey, the service quality and
reliability that drivers perceive can be evaluated; however, only basic results, such as a
simple percentage or degree of satisfaction relative to each criterion could be provided. These
types of results are limited and not really sufficient to represent drivers’ perception of service
quality. They cannot represent appropriately the variability and complexity of human
perception. Another problem with using a survey method is the difficulty of describing the
survey results quantitatively and objectively because surveys primarily use linguistic terms,

35


not exact quantitative scales, to evaluate a subject’s response and those linguistic terms
are determined by subjective human decision making processes. Thusly, there is a need to
interpret quantitatively and objectively drivers’ perceptions as indicated in the results of
the survey, to aggregate the drivers’ responses to various questions related to VMS
service, and to evaluate the overall driver perception of the system. In the ITS Evaluation
Resource Guide, user satisfaction is regarded as one of the measures of effectiveness of
mobility. It is also indicated that user satisfaction measures characterize the difference
between users’ expectations and experience in relation to a service or product. Relative to
this guideline, several evaluations have been performed. Studies of ATIS in Arizona and
Missouri (Orban et al. 2000) were conducted in which an analysis of user satisfaction was
included. In these studies, simple percentages corresponding to each evaluation criterion
were presented as the final results. However, this type of approach does not provide
sufficient or meaningful detail relative to making design decisions when contrary findings
are indicated. This was similar to the methodology and analysis of ATIS. The major
drawback to all of these studies is that they were unable to investigate the individual
differences in the participants’ subjective assessment of service quality. In this study, a
method of evaluating VMS service quality based on fuzzy set theory is introduced. Fuzzy
sets, where a more flexible sense of membership is possible, are classes with “un-sharp”
and vague boundaries. Fuzzy sets theory is a branch of set theory that is useful for the
representation of imprecise knowledge of the type that is prevalent in human concept
formation and reasoning. Its usefulness lays in the concept that fuzzy theory can represent
a type of uncertainty due to vagueness or fuzziness The membership function is the most
important element of the fuzzy approach as it makes it possible for fuzzy set theory to be
used to evaluate uncertain and ambiguous matters. One of the most important and
difficult tasks for applying fuzzy technique is to correctly measure the membership
function. In this research, the role of the membership function is to represent an
individual and subjective human perception as a member of a fuzzy set. This membership
function can represent the degree of the subjective notions of a vague class with an
infinite set of values between 0 and 1. In the procedure describe herein, two membership
functions for five linguistic scales and an evaluation of the relative importance of six
service-related criteria were formulated using results from a survey conducted for this
study. The second fuzzy membership function was determined applying which is
commonly used in multiple decision making analysis. Individual perceptions of the
service were evaluated using results from a previously conducted survey subjected to two
fuzzy membership functions.

2.0 AGGREGATION OF THE INDIVIDUAL PERCEPTIONS

The 322 individual perceptions of VMS service evaluated above should be
aggregated to represent the group’s overall opinion. For aggregating the fuzzy number of
the perceptions, a ”arithmetic mean” of the fuzzy numbers, which represent all individual
perceptions, were calculated using a fuzzy average operation based on the “α-cut”
concept of fuzzy sets and an interval analysis. The outputs from this step are still fuzzy
numbers, and they should be transformed into crisp numbers to be more easily
understood.

36


3.0 DEFUZZIFICATION

To transform the final fuzzy set that represents the group’s overall opinion into crisp
numbers, a defuzzification procedure was conducted. Out of several defuzzification methods,
the defuzzified method developed by Juang, et al. was used due to its simplicity and ease of
computation. It is a mapping model for measuring fuzzy numbers using estimated utility

where
u = the utility to measure or rank a fuzzy number,
AL = the area enclosed to the left of the characteristic function that
characterizes the fuzzy number,
AR = the area enclosed to the right of the characteristic function that
characterizes the fuzzy number,
The utility yields a value between 0 and 1, and the higher utility value indicates the higher
service quality of VMS. Through these procedures, the survey results from the prior study,
which consisted of simple percentages, were converted to an overall measure of service
quality that takes into consideration the variance of human perception and the degree of
importance of the six criteria. The final number represents the overall service quality
perceived by all drivers.

4.0 CONSTRUCTION OF MEMBERSHIP FUNCTIONS

As mentioned previously, the types of fuzzy membership function were determined after
a review of the data. To find the first membership functions for five scales of linguistic
statements, the universe interval, from 0 to 1.0, was partitioned with unit length (0.05)
intervals. Then normalized frequencies of each unit interval were calculated. The shapes of
the histograms derived from these normalized frequencies indicated that a trapezoid
membership function was the most appropriate type of membership function for representing
five scales of linguistic statements indicated that a trapezoidal membership function is
commonly used to represent a fuzzy interval estimate. The trapezoidal membership function
is specified by four parameters {a, b, c, d} as following mentioned in Table 2-2. Finally, to
determine the membership function, three rules for designing the membership function were
considered:
Rule 1: Each membership function overlaps only with the closest neighboring
membership functions
Rule 2: For any possible input data (x), its membership values in all relevant
fuzzy sets should sum to one or nearly so.
Rule 3: The range of top of trapezoid should be approximately matched with the
standard deviation of the input value, x.

Procedures described above and their parameter values. The second set of membership
functions, which represent the fuzzy weight of six criteria, was created by the 23 sets of
weights evaluated. This procedure was repeated with all of the survey results, which created
23 PCMs and sets of weights. Review of the data indicated that triangular fuzzy membership
functions were the most suitable type of membership function for representing the weights of

37


the six criteria. A triangular membership function is specified by three parameters {a, b, c},
and the precise appearance of the function is determined by the choice of parameters

Figure 1.Fuzzy Membership Function of Five Scales Linguistic Statement.

Table 1. The PCM and Weight judged by an Individual Response (Participant i)

These three parameters were finally determined using the min, modal, and max values. The
second set of fuzzy membership functions. As can be seen in the figure, drivers regard
comprehension, accuracy, and usefulness of VMS as being more important than visibility,
legibility to read, and correspondence to their expectance. This result is similar to an earlier
study of criteria for traffic sign design and evaluation. In this study, Dewar conducted a
survey that examined the relative importance of criteria used for traffic sign design and
evaluation and indicated that the criteria related to the content of a traffic sign (e.g.
understandability) were more important than the criteria related to visibility and identification
(e.g. legibility or reaction time).

38


Figure 2. The Fuzzy Weights of Six Criteria

This scale, which represents each level of “agreement” with an integer, was transformed into
a fuzzy number using the first fuzzy membership function. The individual perceptions
transformed as fuzzy sets were aggregated by the fuzzy weighted average based on extended
algebraic operations

Table 2: The Individual Perception and the Arithmetic Mean Value

39


Figure 3: The Final Fuzzy Set Representing the Group’s Perception of VMS Service

5.0 CONCLUSION

In this paper, quality of VMS service was evaluated using the fuzzy approach.
Generally, quality of service is a significant indicator in evaluating the performance of
transportation facilities. However, it is difficult to be measure because the quality of service
that a human perceives is affected by various factors, and it is represented well in qualitative
linguistic terms, but not quantitatively. Current approaches for evaluating the service quality
of transportation facilities are limited because human thinking is subjective and complicated,
and human perception cannot be represented by binary or numerical information. The
proposed method makes numerical evaluation of service quality feasible. To apply this
method, two fuzzy membership functions were determined through a survey. Many previous
studies did not concentrate on the construction of the fuzzy. Membership functions, even
though this is the most significant step for fuzzy applications. In this paper, the first
membership function, which represents five scales of linguistic statements, was constructed
using the interval estimation method. The second set of
fuzzy membership functions, which represent the importance of the weight of six criteria for
evaluating VMS service quality, were determined using eigenvector method. Quality of VMS
service perceived by an individual driver was evaluated using the fuzzy weight average. A set
of 322 quality measures were computed, and they were aggregated and transformed to one
number using the arithmetic fuzzy mean. The defuzzified final value indicates the degree of
satisfaction with VMS service that was perceived by the participating drivers. This value
takes into consideration the variance of human perception and the degree of importance of the
six criteria.

REFERENCES

1. Kulka, J. and V. Novak, Have fuzzy operators a psychological correspondence, Studia
Psychologica, Vol. 26, pp. 131-140, 1984.
2. Labov, W. The Boundaries of Words and their Meanings, In New Ways of Analyzing
Variation in English; Bailey, C. J.; Shuy, R. W., Eds.; Georgetown Univ. Press:
Washington, DC, 1973.

40


3. Lamm, R., E.M. Choueiri, J.C. Hayward, and A. Paluri, Possible Design Procedure to
Promote Design Consistency in Highway Geometric Design on Two-Lane Rural Roads,
Transportation Research Record 1195, TRB, National Research Council, Washington,
D.C., 1988, pp. 111-122.
4. Lamn, R., H. Steffen, and A.K. Guenther, Procedure for detecting Errors in Alignment
Design and Consequences for Safer Design, Transportation Research Record 1445, TRB,
National Research Council, Washington, D.C., 1994, pp.64-72.
5. Lee, D., M. T. Pietrucha, and S. K. Sinha, Application of Fuzzy Logic to Evaluate Driver
Perception of Variable Message Signs. Transportation Research Record 1937 TRB,
National Research Council, Washington, D.C., 2005, pp.96-104.
6. Lee, D., M. T. Pietrucha, and E. T. Donnell, Incorporation of Transportation Experts
Opinions of Median Safety Using a Hierarchical Fuzzy Inference System. In proceeding
of 85 Annual Meeting of the Transportation Research Board, Transportation Research
Board, Washington D.C, 2006
7. Lee, H. T. and S. H. Chen, Using Cpx Index with Fuzzy Numbers to Evaluate Service
Quality, International Transactions in Operational Research Vol. 9 Issue 6 pp 719-
730,2002.
8. Lee, P., K. Lee, and G. Jeon, An Index of Applicability for the Decomposition Method of
Multivariable Fuzzy Systems, IEEE Transactions on Fuzzy Systems, Vol. 3, No. 3, 1995,
pp. 364 – 369.
9. Li, J., K.H. Cheung, W. Rattasiri, S. K. Halgamuge, Self-generating hierarchical fuzzy
systems, Intelligent Sensing and Information Processing, 2004. Proceedings of
International Conference on 2004 pp. 348 – 353.
10. Loizos, A., A Simplified Application of Fuzzy Set Theory for the Evaluation of Pavement
Roughness. Road and Transport Research, Vol. 10, No. 4 2001, pp. 21-32
11. Masuoka, R., N. Watanabe, A. Kawamura, Y. Owada, and K. Asakawa, Neurofuzzy
System- Fuzzy Inference Using a Structured Neural Network, Proceedings of the
International Conference on Fuzzy Logic and Neural Networks at Iizuka, Japan, July 20-
24, 1990, pp. 173-177.
12. Pennsylvania Transportation Institute and Midwest Research Institute, Median Safety
Study (Interstate and Expressways) –Task 5 Report of Survey of Experts, October
2000.Pfefer, R.C., Toward Reflecting Public Perception of Quality of Service in
Planning, Designing, and Operating Highway Facilities. Transportation Research Record
1685, TRB, 1999, pp. 81-89.
13. Pietrucha, Martin T. , Incorporation User Perception into Conventional Engineering
Measures of Effectiveness, Transportation Research Board Conference on Advanced
Modeling Techniques and Quality of Service in Highway Capacity Analysis, July, 2001
14. Ragin, C.C., Fuzzy-Set Social Science, University of Chicago Press, Chicago, 2000.
15. Raju, G.V.S., J. Zhou, and R.A. Kisner, Hierarchical Fuzzy Control, International
Journal of Control, Vol. 54, No. 5, 1991, pp. 1201-1216.
16. Jaydev Mishra and Sharmistha Ghosh, “Normalization In A Fuzzy Relational Database
Model” International journal of Computer Engineering & Technology (IJCET), Volume
3, Issue 2, 2012, pp. 506 - 517, Published by IAEME
17. Anand Handa and Ganesh Wayal, “Software Quality Enhancement Using Fuzzy Logic
With Object Oriented Metrics In Design” International journal of Computer Engineering
& Technology (IJCET), Volume 3, Issue 1, 2012, pp. 169 - 179, Published by IAEME

41

Governance of fuzzy systems to predict driver perception of service quality

More Related Content

What's hot (14)

Viewers also liked (16)

Similar to Governance of fuzzy systems to predict driver perception of service quality (20)

More from iaemedu (20)

Governance of fuzzy systems to predict driver perception of service quality